Stop and return the error code when update_datapath fails.
update_datapath prepares resources for the vdpa device.
The driver should not perform any further actions
if update_datapath returns an error.
Fixes: a3f8150eac6d ("net/ifcvf: add ifcvf vDPA driver")
Cc: sta...@dpdk.org
Signed-off-by: Tae
On 2022-11-04 13:03, Morten Brørup wrote:
When built with stats enabled (RTE_LIBRTE_MEMPOOL_STATS defined), the
performance of mempools with caches is improved as follows.
When accessing objects in the mempool, either the put_bulk and put_objs or
the get_success_bulk and get_success_objs statist
On 2022-11-04 11:01, Morten Brørup wrote:
From: Mattias Rönnblom [mailto:hof...@lysator.liu.se]
Sent: Friday, 4 November 2022 09.59
On 2022-11-03 09:59, Morten Brørup wrote:
From: Mattias Rönnblom [mailto:hof...@lysator.liu.se]
Sent: Wednesday, 2 November 2022 18.53
On 2022-11-02 10:09, Morten
> -Original Message-
> From: Andrew Rybchenko
> Sent: 2022年11月6日 18:33
> To: Ye, MingjinX ; lihuisong (C)
> ; dev@dpdk.org
> Cc: sta...@dpdk.org; Zhou, YidingX ; Singh, Aman
> Deep ; Zhang, Yuying
>
> Subject: Re: [PATCH v4 1/2] app/testpmd: fix vlan offload of rxq
>
> On 11/4/22 14:33
These newly added files for Congestion Management look in-line to
Traffic Metering part,
would like @Cristian to also have a look.
Regards
Aman
On 11/6/2022 3:38 PM, Andrew Rybchenko wrote:
@Aman, @Yuying, please, help to review the patch.
On 10/12/22 12:01, Sunil Kumar Kori wrote:
Please re
For Arm aarch, power monitor uses WFE instruction to enable, which can
not exit automatically within the time limit. This means
'rte_power_monitor_wakeup' API needs to be called to wake up sleep cores
if there is no store operation to monitored address.
Furthermore, we disable power monitor featur
For Arm aarch, use WFE instruction to enable power monitor API, and use
SEV instruction to enable wake up API.
Signed-off-by: Feifei Wang
Reviewed-by: Ruifeng Wang
---
lib/eal/arm/include/rte_pause_64.h | 5 ++-
lib/eal/arm/rte_cpuflags.c | 5 +++
lib/eal/arm/rte_power_intrinsics.c |
For wait scheme generic helper, add 8 bits case.
Signed-off-by: Feifei Wang
Reviewed-by: Ruifeng Wang
---
lib/eal/arm/include/rte_pause_64.h | 27 ---
1 file changed, 24 insertions(+), 3 deletions(-)
diff --git a/lib/eal/arm/include/rte_pause_64.h
b/lib/eal/arm/include
For Arm aarch, use WFE instructions to enable PMD power management.
Test Results:
dynamic instructions over 1sec without wfe with wfepercentage
ampere-altra6,298,483,712 9,117,624 -99.855%
thunderx2 6,990,909,373 3,247,226 -99.
Hi Andrew,
> -Original Message-
> From: Andrew Rybchenko
> Sent: Sunday, November 6, 2022 5:58 PM
> To: Wang, YuanX ; Singh, Aman Deep
> ; Zhang, Yuying
> Cc: dev@dpdk.org; Ding, Xuan ; Tang, Yaqi
>
> Subject: Re: [PATCH v2] app/testpmd: fix protocol headers display for Rx
> buffer spli
On 11/7/22 05:18, Huisong Li wrote:
Some capabilities (like, rx_offload_capa and tx_offload_capa) of bonding
device in dev_info is zero when no slave is added. And its capability will
be updated when add a new slave device.
The capability to update dynamically may introduce some problems if not
Some of the HW has support for choosing memory pools based on
the packet's size. The pool sort capability allows PMD/NIC to
choose a memory pool based on the packet's length.
On multiple mempool support enabled, populate mempool array
accordingly. Also, print pool name on which packet is received.
> On 11/4/22 04:08, Hernan Vargas wrote:
> > Add function to detect if de-ratematch pre-processing is recommended for
> > SW corner cases.
> > Some specific 5GUL FEC corner cases may cause unintended back pressure
> > and in some cases a potential stability issue on the ACC100.
> > The PMD can dete
> Acked-by: Kai Ji
> > Subject: [PATCH] crypto/ipsec-mb: fix qp setup for secondary process
> >
> > If a secondary process is using a queue pair that has been setup by the
> > primary
> > process, we need to reset mb_mgr pointers.
> > This commit removes an error return in this case, allowing sec
Some capabilities (like, rx_offload_capa and tx_offload_capa) of bonding
device in dev_info is zero when no slave is added. And its capability will
be updated when add a new slave device.
The capability to update dynamically may introduce some problems if not
handled properly. For example, the rec
Some capabilities (like, rx_offload_capa and tx_offload_capa) of bonding
device in dev_info is zero when no slave is added. And its capability will
be updated when add a new slave device.
The capability to update dynamically may introduce some problems if not
handled properly. For example, the rec
BR
Rongwei
> -Original Message-
> From: Andrew Rybchenko
> Sent: Sunday, November 6, 2022 18:03
> To: Rongwei Liu ; Matan Azrad ;
> Slava Ovsiienko ; Ori Kam ;
> NBU-Contact-Thomas Monjalon (EXTERNAL) ; Aman
> Singh ; Yuying Zhang
> ; Ferruh Yigit
> Cc: dev@dpdk.org; Raslan Darawsheh
Hi,
> -Original Message-
> From: Spike Du
> Sent: Wednesday, November 2, 2022 1:44 PM
> To: Matan Azrad ; Slava Ovsiienko
> ; Ori Kam ; NBU-Contact-
> Thomas Monjalon (EXTERNAL) ; Shahaf Shuler
>
> Cc: dev@dpdk.org; Raslan Darawsheh ;
> sta...@dpdk.org
> Subject: [PATCH v2] mlx5/testpmd:
Hi,
> -Original Message-
> From: Alexander Kozyrev
> Sent: Wednesday, November 2, 2022 4:25 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava
> Ovsiienko ; Matan Azrad ;
> Xueming(Steven) Li
> Subject: [PATCH] net/mlx5: fix shared Rx queue config reuse
>
> There is a
From: Spike Du
> When testpmd quit with mlx5 avail_thresh enabled, a rte timer handler
> delays to reconfigure rx queue to re-arm this event. However at the same
> time, testpmd is destroying rx queues.
> It's never a valid use case for mlx5 avail_thresh. Before testpmd quit, user
> should disa
> There is a check for the configuration match between all the Rx queues
> shared among multiple ports in DPDK.
> This check ensures that the configuration is the same.
>
> The issue is this check takes place before the queue is released and
> configured again in case of reconfiguration.
> That
> From: Andrew Rybchenko [mailto:andrew.rybche...@oktetlabs.ru]
> Sent: Sunday, 6 November 2022 13.00
>
> On 11/6/22 14:50, Morten Brørup wrote:
> >> From: Andrew Rybchenko [mailto:andrew.rybche...@oktetlabs.ru]
> >> Sent: Sunday, 6 November 2022 12.41
> >>
> >> On 11/4/22 15:03, Morten Brørup wro
On 11/6/22 14:50, Morten Brørup wrote:
From: Andrew Rybchenko [mailto:andrew.rybche...@oktetlabs.ru]
Sent: Sunday, 6 November 2022 12.41
On 11/4/22 15:03, Morten Brørup wrote:
[...]
+/**
+ * @internal When stats is enabled, store some statistics.
+ *
+ * @param cache
+ * Pointer to the mem
> From: Andrew Rybchenko [mailto:andrew.rybche...@oktetlabs.ru]
> Sent: Sunday, 6 November 2022 12.41
>
> On 11/4/22 15:03, Morten Brørup wrote:
[...]
> > +/**
> > + * @internal When stats is enabled, store some statistics.
> > + *
> > + * @param cache
> > + * Pointer to the memory pool cache.
On 11/4/22 15:03, Morten Brørup wrote:
When built with stats enabled (RTE_LIBRTE_MEMPOOL_STATS defined), the
performance of mempools with caches is improved as follows.
When accessing objects in the mempool, either the put_bulk and put_objs or
the get_success_bulk and get_success_objs statistics
On 11/4/22 15:03, Morten Brørup wrote:
This patch adds statistics for unregistered non-EAL threads, which was
previously not included in the statistics.
Add one more entry to the stats array, and use the last index for
unregistered non-EAL threads.
The unregistered non-EAL thread statistics are
On 11/4/22 15:03, Morten Brørup wrote:
Split stats from debug, to make mempool statistics available without the
performance cost of continuously validating the debug cookies in the
mempool elements.
mempool_perf_autotest shows the following improvements in rate_persec.
The cost of enabling memp
Hi,
> -Original Message-
> From: Suanming Mou
> Sent: Sunday, November 6, 2022 5:26 AM
> To: Matan Azrad ; Slava Ovsiienko
>
> Cc: dev@dpdk.org; Raslan Darawsheh
> Subject: [PATCH v2] net/mlx5: fix Windows flow table and queue routine
>
> The macro HAVE_MLX5_HWS_SUPPORT was introduced
Hi,
> -Original Message-
> From: Shun Hao
> Sent: Thursday, November 3, 2022 5:33 PM
> To: Slava Ovsiienko ; Matan Azrad
> ; Ori Kam ; Suanming Mou
>
> Cc: dev@dpdk.org; Raslan Darawsheh ;
> sta...@dpdk.org
> Subject: [PATCH v1] net/mlx5: fix incorrect match mask for meter flow
>
> Ther
Hi,
> -Original Message-
> From: Gregory Etelson
> Sent: Thursday, November 3, 2022 12:44 PM
> To: dev@dpdk.org
> Cc: Gregory Etelson ; Matan Azrad
> ; Raslan Darawsheh ;
> sta...@dpdk.org; Slava Ovsiienko ; Anatoly
> Burakov ; Dmitry Kozlyuk
>
> Subject: [PATCH] common/mlx5: fix shared
Hi,
> -Original Message-
> From: Dariusz Sosnowski
> Sent: Thursday, November 3, 2022 12:10 PM
> To: Matan Azrad ; Slava Ovsiienko
>
> Cc: dev@dpdk.org; Raslan Darawsheh
> Subject: [PATCH] net/mlx5: forbid direction attributes in transfer flow rules
>
> Since [1] flow API forbids usage
On 10/19/22 06:32, Huisong Li wrote:
Currently, by default, bond4 will first try to enable allmulti and
then enable promiscuous if fail to enable allmulti. On reception,
whether unicast and multicast packets should be dropped depends on
which mode has been enabled on the bonding interface.
In fa
On 11/6/22 13:46, Andrew Rybchenko wrote:
On 10/10/22 12:35, Nathan Skrzypczak wrote:
This patch changes the advertised link speed for the memif
driver from 10G to 100G as the memory interfaces can reach
higher throughputs than 10G with large packets.
Signed-off-by: Nathan Skrzypczak
Acked-b
On 10/14/22 20:23, Stephen Hemminger wrote:
The original behavior of testpmd was to kill itself when
it received a SIGINT or SIGTERM. This makes it hard to use
testpmd in test automation where forwarding loop is started
and then stopped via SIGTERM.
Can automatic stop it using SIGINT?
Signed
On 10/14/22 20:23, Stephen Hemminger wrote:
Since f_quit is set in a signal handler it needs
to be marked as volatile. Otherwise, compler is allowed
compler -> compiler
to optimize away access to it.
Signed-off-by: Stephen Hemminger
Reviewed-by: Andrew Rybchenko
I need non-RFC version t
On 10/10/22 12:35, Nathan Skrzypczak wrote:
This patch changes the advertised link speed for the memif
driver from 10G to 100G as the memory interfaces can reach
higher throughputs than 10G with large packets.
Signed-off-by: Nathan Skrzypczak
Acked-by: Andrew Rybchenko
On 10/19/22 16:13, Bruce Richardson wrote:
On Wed, Sep 21, 2022 at 04:26:11PM +0200, Thomas Monjalon wrote:
I think this patch requires a techboard vote
as it is dropping some user-facing naming.
I think a better solution is to centralize the use of aliases. I've just
posted a patch to this th
On 11/4/22 14:33, Ye, MingjinX wrote:
-Original Message-
From: lihuisong (C)
Sent: 2022年11月4日 18:18
To: Ye, MingjinX ; dev@dpdk.org
Cc: sta...@dpdk.org; Zhou, YidingX ; Singh, Aman
Deep ; Zhang, Yuying
Subject: Re: [PATCH v4 1/2] app/testpmd: fix vlan offload of rxq
在 2022/11/4 16:
@Aman, @Yuying, please, help to review the patch.
On 10/12/22 12:01, Sunil Kumar Kori wrote:
Please review the following changes and provide feedback.
Regards
Sunil Kumar Kori
-Original Message-
From: sk...@marvell.com
Sent: Thursday, September 29, 2022 3:25 PM
To: Aman Singh ; Yuyin
On 11/1/22 06:41, humin (Q) wrote:
Reviewed-by: Min Hu (Connor)
在 2022/10/29 11:50, Huisong Li 写道:
Some capabilities (like, rx_offload_capa and tx_offload_capa) of bonding
device in dev_info is zero when no slave is added. And its capability
will
be updated when add a new slave device.
The
On 10/4/22 11:31, Andrew Rybchenko wrote:
On 9/28/22 12:24, Rongwei Liu wrote:
The transfer domain rule is able to match traffic wire/vf
origin and it means two directions' underlayer resource.
In customer deployments, they usually match only one direction
traffic in single flow table: either f
On 10/18/22 17:50, Yuan Wang wrote:
The "show config rxhdrs" cmd displays the configured protocol headers
that are used for protocol-based buffer split.
However, it shows "inner-ipv6" as "inner-ipv4".
This patch fixes that by adjusting the order of condition judgments.
Fixes: 52e2e7edcf48 ("app
On 10/27/22 07:05, lihuisong (C) wrote:
在 2022/10/26 19:07, Gregory Etelson 写道:
MLX5 SR-IOV TX engine will not transmit Ethernet frame
if destination MAC address matched local port address. The frame ether
looped-back to RX or dropped, depending on the port configuration.
Application running o
On 11/1/22 05:34, humin (Q) wrote:
在 2022/10/28 10:36, Huisong Li 写道:
Normally, the Rx/Tx offload capability of bonding interface is
the intersection of the capability of all slave devices. And
Rx/Tx offloads configuration of slave device comes from bonding
interface. But now there is a risk tha
On 11/2/22 10:07, Li, WeiyuanX wrote:
-Original Message-
From: humin (Q)
Sent: Tuesday, November 1, 2022 10:21 AM
To: Ivan Malov ; dev@dpdk.org
Cc: Li, WeiyuanX ; Chas Williams ;
Hari Kumar Vemula ; sta...@dpdk.org;
Andrew Rybchenko
Subject: Re: [PATCH] net/bonding: set initial value of
On 11/2/22 06:35, Yuan, DukaiX wrote:
-Original Message-
From: Ivan Malov
Sent: 2022年11月2日 0:19
To: dev@dpdk.org
Cc: Jiang, YuX ; Chas Williams ; Min Hu (Connor)
; Burakov, Anatoly ; sta...@dpdk.org; Andrew
Rybchenko
Subject: [PATCH] net/bonding: fix device configure reentrancy
As pe
On 11/3/22 17:00, David Marchand wrote:
With symbols going though experimental/stable stages, we accumulated
a lot of discrepancies about inclusion of the rte_compat.h header.
Some headers are including it where unneeded, while others rely on
implicit inclusion.
Fix unneeded inclusions:
$ git g
On 11/1/22 22:32, Stephen Hemminger wrote:
On Tue, 1 Nov 2022 19:18:53 +0300
Ivan Malov wrote:
diff --git a/drivers/net/bonding/eth_bond_private.h
b/drivers/net/bonding/eth_bond_private.h
index d067ea8c9a..7171516d0d 100644
--- a/drivers/net/bonding/eth_bond_private.h
+++ b/drivers/net/bondi
On 11/2/22 04:23, Chaoyong He wrote:
Fix the check logic of the index of the array, which
caused the out of bounds write problem.
Coverity issue: 381616
Fixes: 0666af498761 ("net/nfp: support RSS based on VXLAN inner layer")
Signed-off-by: Chaoyong He
Reviewed-by: Niklas Söderlund
Applied t
On 11/3/22 16:49, Andrew Boyer wrote:
These patches will make no functional difference, but should
eliminate four coverity issues.
Alternatively, we could just mark them as False Positive or
Ignored in coverity.
Signed-off-by: Andrew Boyer
Andrew Boyer (2):
net/ionic: fix up minor coverity
> From: Morten Brørup
> Sent: Saturday, 5 November 2022 14.19
>
> Zero-copy access to the mempool cache is beneficial for PMD
> performance, and must be provided by the mempool library to fix [Bug
> 1052] without a performance regression.
>
> [Bug 1052]: https://bugs.dpdk.org/show_bug.cgi?id=1052
51 matches
Mail list logo
